Deep complex convolutional network for fast reconstruction of 3D late gadolinium enhancement cardiac MRI. (30th April 2020)

• Record Type: Journal Article
• Authors: El‐Rewaidy, Hossam; Neisius, Ulf; Mancio, Jennifer; Kucukseymen, Selcuk; Rodriguez, Jennifer; Paskavitz, Amanda; Menze, Bjoern; Nezafat, Reza
• Is Part Of: NMR in biomedicine. Volume 33:Number 7(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-04-30
• Subjects: complex convolutional network -- deep learning -- image reconstruction -- late gadolinium enhancement -- MRI; Nuclear magnetic resonance -- Periodicals; Magnetic Resonance Spectroscopy -- Periodicals; 574
• DOI: 10.1002/nbm.4312
• Languages: English
• ISSNs: 0952-3480

Abstract

Abstract : Several deep‐learning models have been proposed to shorten MRI scan time. Prior deep‐learning models that utilize real‐valued kernels have limited capability to learn rich representations of complex MRI data. In this work, we utilize a complex‐valued convolutional network ( ℂ Net) for fast reconstruction of highly under‐sampled MRI data and evaluate its ability to rapidly reconstruct 3D late gadolinium enhancement (LGE) data. ℂ Net preserves the complex nature and optimal combination of real and imaginary components of MRI data throughout the reconstruction process by utilizing complex‐valued convolution, novel radial batch normalization, and complex activation function layers in a U‐Net architecture. A prospectively under‐sampled 3D LGE cardiac MRI dataset of 219 patients (17 003 images) at acceleration rates R = 3 through R = 5 was used to evaluate ℂ Net. The dataset was further retrospectively under‐sampled to a maximum of R = 8 to simulate higher acceleration rates. We created three reconstructions of the 3D LGE dataset using (1) ℂ Net, (2) a compressed‐sensing‐based low‐dimensional‐structure self‐learning and thresholding algorithm (LOST), and (3) a real‐valued U‐Net (realNet) with the same number of parameters as ℂ Net. LOST‐reconstructed data were considered the reference for training and evaluation of all models. The reconstructed images were quantitatively evaluated using mean‐squared error (MSE) and the structural similarity index measure (SSIM), and subjectively evaluated by three independent readers. Quantitatively, ℂ Net‐reconstructed images had significantly improved MSE and SSIM values compared with realNet (MSE, 0.077 versus 0.091; SSIM, 0.876 versus 0.733, respectively; p < 0.01). Subjective quality assessment showed that ℂ Net‐reconstructed image quality was similar to that of compressed sensing and significantly better than that of realNet. ℂ Net reconstruction was also more than 300 times faster than compressed sensing. Retrospective under‐sampled images demonstrate the potential of ℂ Net at higher acceleration rates. ℂ Net enables fast reconstruction of highly accelerated 3D MRI with superior performance to real‐valued networks, and achieves faster reconstruction than compressed sensing. Abstract : A complex‐valued convolutional neural network (ℂNet) utilizes complex convolutional layers, novel radial batch normalization, and complex ReLU in U‐net architecture for fast reconstruction of highly under‐sampled 3D cardiac MR data. A large dataset of 17003 cardiac late gadolinium enhancement MR images was used for training and evaluating ℂNet. ℂNet achieved more than 300‐fold of acceleration in reconstruction time than compressed sensing methods and outperformed the conventional real‐valued networks using quantitative and qualitative measures.